Evidence-Based Management: The “What” and “Why”?
Description of the principles of evidence-based management, and how it underpins organisational performance
EVIDENCE BASED MANAGEMENT
7/12/20269 min read
MEric Barends, Denise M. Rousseau and Rob B. Briner at the Center for Evidence-Based Management describe the problem plainly: most management decisions are not based on the best available evidence. Instead, practitioners rely on personal experience, the opinions of management gurus, and the benchmarked practices of competitor organisations — sources of information that are consistently less reliable than they appear. The result is billions of dollars spent on management practices that are ineffective, or worse, actively harmful to the organisations that adopt them.
The Problem With Gut Feeling
Every manager makes decisions. Most of them, most of the time, are made quickly on the basis of experience, precedent, or instinct. For routine, lower-stakes questions, this is entirely appropriate. The human capacity for pattern recognition is strong, and an experienced manager usually has a reasonable feel for operational matters they have dealt with many times before.
The trouble arises when the same fast, intuitive approach is applied to complex, high-stakes decisions where the situation is genuinely novel, the evidence is available but ignored, and the costs of getting it wrong are significant. Research on cognitive bias tells us that the human mind is systematically prone to such errors and, in these cases, experience can actually deepen those errors rather than reduce them: Kahneman, Bazerman, and a substantial body of psychological research document how cognitive shortcuts that work well in simple, familiar situations produce poor outcomes in environments of genuine complexity and uncertainty.
The management field has responded slowly to these findings. Medicine responded faster.
An Idea From Medicine
Evidence-based medicine emerged in the early 1990s through the work of David Sackett and colleagues at McMaster University, building on foundations laid by epidemiologist Archie Cochrane. Their argument was direct: clinical decisions should integrate the best available scientific research with individual clinical expertise and the unique values and circumstances of each patient. Sackett and colleagues coined the phrase "the conscientious, explicit and judicious use of current best evidence". It was not saying that science should override clinical judgment, nor that patient preferences were secondary. It was saying that decisions are better all round for using all of these sources.
The diagnostic power of this approach becomes clearer when we think about how a hospital treats a complex patient. A physician who faces a complex case will draw on research evidence, such as published meta-analyses on treatment efficacy. However, they also draw on years of clinical experience that helps contextualise those findings for the specific patient in front of them. They take on board what the patient values most: peace-of-mind, pain levels, long-term prognosis etc. Local considerations might also play a role, such as the patient’s home environment or services available at the hospital. No single source dominates. All inform the ultimate course of action.
Evidence-based management applies the same logic to management decisions. The definition offered by Barends, Rousseau and Briner at the Center for Evidence-Based Management strongly reflects Sackett's medical description.
Four Sources of Evidence
The first principle of evidence based management is to use four diverse sources of information:
"Scientific": Findings from published research in management, organisational behaviour, economics, psychology, and adjacent disciplines. The volume of this research has grown substantially over recent decades on how organisations create or destroy value. When properly appraised, this evidence provides something that individual experience cannot: findings drawn from large samples, across diverse contexts, with systematic controls for alternative explanations, and objective interpretation.
“Local”: This refers to local context such as organisational evidence - data, facts and figures gathered from within the organisation itself or from the “patient”. In a business context, it can include financial metrics, customer satisfaction scores, staff retention rates and operational efficiency. Organisational data is essential for identifying the status of a situation, whether a problem exists, and what might be causing it.
"Practitioner": This refers to the professional judgment accumulated by practitioners through reflection on repeated experience in similar situations. It is not the same as personal opinion or intuition. It is the tacit, professional knowledge of a subject expert, such as a functional marketing, operational or technology leader. Experiential evidence is particularly valuable for assessing whether research findings are applicable in a particular context, whether an intervention that worked elsewhere is likely to work here, and whether organisational data can be trusted.
"Stakeholder": This represents the values, concerns, and interests of the individuals and groups who are affected by a decision and its consequences: Employees, customers, shareholders, and the wider public will all bring perspectives that shape how decisions will be received and whether the conditions for successful implementation actually exist. Organisations that serve different stakeholders can arrive at different decisions about a similar sounding issue because, for instance, they serve different customer segments.
Evidence-based management approaches draw on all four of these sources. It is an integrative and critical endeavour.
A Systematic Process
Someone leading an evidence based management process then needs a structure for turning evidence into a decision. The Center for Evidence-Based Management describes one such process with six steps, sometimes called the 6As, which operationalise the evidence-based approach in practice. (There are others, but this works well).
Ask: The process begins with translating a practical problem into a clear, answerable question. This discipline alone is more demanding than it sounds. Managers are frequently tempted to leap to solutions before the problem has been precisely defined. A vague question produces vague evidence searches and ultimately poor decisions. "We have a staff retention problem, so need to fix it" is weaker than "We need to understand the drivers of voluntary exits among technical staff in the first 18 months of employment? And which interventions have demonstrated effectiveness in comparable contexts to our’s?. The latter can be better researched and tested.
Acquiring: From a clear question, the practitioner moves to systematically finding and retrieving relevant evidence across all four source types. Each project will be different, but this step is likely to include some amount of searching academic databases, pulling and analysing organisational data, consulting colleagues, and engaging stakeholders. It is a deliberate, structured search where evidence requests are grounded in the “Ask”.
Appraising: The third step involves critically evaluating the trustworthiness and relevance of each piece of evidence. Most evidence is imperfect, which means we need to appreciate and factor in the limitations. Academic studies vary in methodological rigour. Organisational data may be inaccurate or incomplete. Experiential accounts may be narrow from any one person. Stakeholder views may be tainted by interest.
Aggregating: The fourth step is about weighing and synthesising the different evidence sources collectively. This step looks for where evidence points in a consistent direction, where sources are in conflict. and how much confidence is warranted in different conclusions.
Applying: The fifth step uses aggregated evidence to inform a decision and its implementation.
Assessing: The sixth step involves evaluating the outcome of the decision once implemented. This step closes the loop, generating new experiential and organisational evidence that improves the quality of future decisions. The same discipline needs to be retained.
When Is EBMgmt Needed?
No serious argument exists that evidence-based management should be applied, in full, for all decisions. Life would grind to a halt. We make thousands of decisions every day. The overwhelming majority of such decisions are handled perfectly well through intuition, knowledge of how similar situations have played out in the past, a question to a knowledgeable colleague, or straightforward reasoning. The cost of a formal approach would exceed the benefit.
We need to recognise when an evidence-based approach merits the investment of time and effort. Several factors are at play:
Complexity: When a decision involves many interacting types of information or variables, where the causal relationships are unclear, and where the system under consideration may behave differently in response to the intervention than simple reasoning would predict. Organisational change, culture transformation, technological disruption and market entry are examples that fall into this category - often drawing from a combination of (e.g.) organisational data, employee and manager input, and vendor insight.
Stakes: The higher the cost of a wrong decision in financial, reputational or human terms, then the more the investment in evidence is justified. A low-stakes decision might incur a small cost. A strategic bet that fails may take years to repair.
Correctability: Some decisions can be reversed or adjusted cheaply and quickly if they turn out to be wrong. Others lock in to pathways that are difficult and costly to undo. Irreversible decisions demand a higher standard of evidence-based management.
Availability: For some decisions, good, relevant evidence simply does not yet exist, like when entering new markets or when disruptive technologies emerge. The appropriate approach could then be to generate scenarios based on the evidence, and treat the decision as a series of hypotheses to be tested that generate necessary evidence.
A final point about "Speed" i.e the urgency of the decision, versus the time to take an evidence-based approach: An ever growing concern is the speed at which data and analysis can be sourced and used. The volume of market intelligence, competitive information, academic research, and organisational data that circulates through large organisations has grown dramatically. The problem today is often less about finding information than about turning it into better decisions at speed. This is a different class of problem, which may require work on the infrastructure and technologies deployed to support evidence based management.
The Case for the Approach
A fair challenge to evidence-based management is the observation that its effectiveness has not itself been demonstrated through the kind of rigorous scientific study it advocates. There is no large-scale, randomised controlled trial comparing organisational outcomes for managers trained in evidence-based methods against a comparable control group. It would be hard to do, but nonetheless, it's a limitation that proponents acknowledge openly.
What exists instead is a substantial set of indicators to show that knowledge and critical thinking produce better outcomes. Evidence-based management offers an architecture that subsumes the reasons that these related practices deliver value:
Data-driven decisions: McKinsey's research consistently finds that organisations which embed data systematically into decision-making demonstrate EBITDA improvements of up to 25% compared to peers. Their DataMatics research, drawing on surveys of over 800 companies, found that intensive users of customer analytics are 23 times more likely to clearly outperform competitors in new customer acquisition, and 19 times more likely to be profitable. Forrester Consulting, in a separate body of work, found that companies using data tools for decision-making are 58% more likely to meet revenue goals and 162% more likely to surpass them.
R&D: Similarly, firms with high R&D intensity show consistently higher stock market valuations and future operating performance than their less research-intensive peers, a pattern documented across multiple decades and market contexts.
Intellectual/Knowledge capital: The intellectual capital literature shows that knowledge-intensive firms consistently find that the depth of their accumulated expertise and their capacity to learn systematically from experience are among the most durable sources of competitive advantage
Agile and test-and-learn approaches (ideation, pilot testing, and rapid iteration) mirrors the principles of the Assessing step in evidence-based management. Organisations that embed this discipline find that they reduce the cost of failure, accelerate the rate of improvement, and build capabilities that sustain new sources of value creation.
Forecasting: In the field of forecasting, Philip Tetlock's research, particularly the Good Judgment Project conducted with the intelligence research agency IARPA, recruited thousands of volunteer forecasters to predict geopolitical and economic events in competition with one another. The finding that generated most attention was the existence of a small group of "superforecasters" who consistently outperformed professional intelligence analysts with access to classified information by margins that were, as Tetlock's team described them, "stunning." The superforecasters were not the most academically credentialled participants, nor the deepest subject-matter experts. What distinguished them was a particular set of habits: they gathered evidence from multiple and varied sources, they thought probabilistically rather than in binary terms, and updated their views in increments when new information arrived. They actively sought out evidence that might contradict their current beliefs, and they tracked their past predictions and learned from their mistakes. In short, they behaved like evidence-based practitioners. The superforecaster approach beat the wisdom of the crowd by 60% across the first two years of the tournament.
A Critical Mindset, Not Just a Process
Evidence-based management is sometimes misunderstood as a demand for certainty before action, or as an argument that numbers always trump judgment, or as a technique that takes more time than busy managers have available. None of these characterisations is accurate.
The approach does not expect to find conclusive evidence. Evidence is rarely conclusive.
It does not replace professional judgment - if anything, it nurtures and amplifies it - by providing a vehicle to capture and use it to best effect.
The approach does not demand extensive time. Many of the most important evidence-based disciplines can be embedded in existing decision-making processes without much additional overhead. It competes favourably when considering the time cost of ignoring evidence that could have mitigated risks.
What evidence-based management does need is a cultural commitment from those participating, such that it becomes normal and safe to ask questions like, "what is the evidence for that?" and “doesn’t this data contradict this other data?” Views are held more lightly. But commitment to the process is held more tightly.
This is all adoptable and learnable, and benefits achievable that surprisingly few companies derive.
This article draws on Evidence-Based Management: The Basic Principles by Eric Barends, Denise M. Rousseau and Rob B. Briner (Center for Evidence-Based Management, 2014), available at cebma.org; Philip Tetlock and Dan Gardner, Superforecasting: The Art and Science of Prediction (2015); McKinsey Global Institute research on data-driven organisations; and published academic literature on evidence-based medicine, cognitive bias, and organisational decision-making.